Unravelling the 3D thermal environment differences between forest center and edge: A case study on 22 urban forests in Hefei city, China

Abstract

Urban forests with various structures can bring considerable but divergent biophysical cooling and humidification effects on their local climate. Thus, it is crucial to unravel the 3D thermal environment within urban forests and their relationship with forest structure, which are helpful for the urban forest planning and design. In this study, we continuously observed the air temperature (T_a) at different vertical layers from canopy to land surface as well as the soil surface temperature (T_s) from the forest center to 5 m outside the forest edge across 22 urban forests in Hefei city, China. Indicators of forest structure such as tree height, diameter at breast height (DBH), crown diameter and leaf traits were associated with their 3D thermal environments for exploring the underlying mechanisms. We found that T_s was 1.43 °C lower than the understory air temperature (T_{a understory}) in forest center but 10.90 °C higher than T_{a understory} outside the forest. Additionally, tree height largely influenced the buffering distance from forests center to the places with T_s = T_{a understory} (L_{Ts=Ta understory Lcenter}), being 4.41 m, 5.80 m and 7.75 m in short (< 7 m), medium (7–9 m) and tall (>9 m) canopy forests, respectively. The temperature difference between forest center and 5 m outside the forest (ΔTemperature) varied significantly at different vertical layers, with ΔT_s greater than 10 °C, ΔT_{a understory} and ΔT_{a bottom canopy} at around 2 °C, and no difference for ΔT_{a upper canopy}. Regression analysis showed different relationships of forest structure and leaf traits with ΔTemperature between vertical layers. Tree height, forest area and DBH showed significant positive relationships with L_{Ts=Ta understory Lcenter}. The study, for the first time, demonstrate the 3D thermal environments of urban forests, quantify the role of forest structure and leaf traits in predicting forest cooling.